• Aucun résultat trouvé

Off-Line Program Tes'ting

N/A
N/A
Protected

Academic year: 2022

Partager "Off-Line Program Tes'ting "

Copied!
13
0
0

Texte intégral

(1)

• •

UNIVAC·

S O L I D - S T A T E

90

Off-Line Program Tes'ting

• •• •

• • •

<

• •

• •

• •

• •

• •

fl. --~

.,

. ...

"

....

-. -

•..

~

.. . .

'"

.-

(2)

CONTENTS

PURPOSE. • • • "

. . . . . . . . . .

CONVENTIONS. •

. . . . . . . . . . . . . . . . . . . . .

PROCEDURES • •

. . . . . . . . . . . . . . . .

Verify Key Punching. • •

. . . . . . . . . . .

Verify Legality of "a" Addresses In All Program Cards ••

Verify Legality of Operation Codes. • • • • • • • • • • Verify Legality of .~ .. Addresses In Instruction Cards.

Verify Legality of "c" Addresses In Instruction Cards ••

Verify Continuity of "c" Address to "a" Address.

. . . .

· . .

· . .

· . .

· . .

I I 5 5 6 6 7 8 9

Verify Insertion of 22 Tests of HSR Buffer • • • • • • • • •• 9 CONSIDERATIONS • • • • • • • • • • • • • • • • • • • • • • 10

(3)

I. PURPOSE

A. To suggest methods of analyzing the clerical accuracy of machine coded problems for the UNIVAC Solid-State 90 using conventional Punched-Card equipment prior to program testing on the computer.

B. To establish standards to aid in program testing at the computer and to assure efficient use of time at program testing facilities.

II. CONVENTIONS

A. To aid in developing a uniform testing procedure, the illustrated Coding Chart (U1711), Memory Chart (U1710) and Program Card form are recommended.

Coding Chart Card

Field Name Col umns Purpose

A Page No. 11 - 13 A sequential number to identify the page of coding.

B Line No. 14 - 15 To identify the line of coding within the page.

C Suffix 16 Optional field. May be used

to designate an insertion.

D Instruction 17 - 20 Designates the memory location

Location where the instruction/constant

( a) is to be stored.

E Operation Code 21 - 22 Designates the machine opera- tion to be performed. (2 MSD of a constant)

F Data Location 23 - 26 Designates the memory location

(m) of the data to be handled. (4

next MSD of a constant)

G Next Instruction 27 - 30 Designates the memory location ( c) of the next instruction to be

executed. (4 LSD of a constant)

H. Key 31 Designates how the instruction/

constant is to be stored.

1

(4)

'" .... __

~_~

__ e

~ SOLlD·STATE COMPUTER· CODING CHART ~6wa.~~ TRAININQ D . . . ARTIII.NT DIVISION Of SPUIY R ... ND CORPOR ... ·lION

ICARD I ROUTINE I

1111 IV

I

" . I _ . _ ' " ' 1 _ _ .'" - r-n"'\0:11""'m_ -. ,..~'"~;

REFERENCE INSTRUCTION ~ K

NOTES

SEQUENCE LOCATION WORD E TIME

PAGK ILINE , X a Op J

m

I c I--y rA rX rL

, I , I I I I I I

VI , , , 1 I I I I I I I I I I I I I I I I I I

I I , I , I , ,

, . t - -

1vII

2 I , , , I I I I I I I I I I I , I I I I I I I I , I I

f - - I

, I I I I

3 , I I I I I I I I I

I I I I I

I I-- I

I I I

4 L , I "- I L. I I

5

I. THE NAME OF THE APPLICATION 6

II. THE NAME OF THE PROGRAM

7 I

8 III. CARD NUMB ER (for use in assembly systp.ms)

9 IV. ROUTINE DESIGN ATION (for use in assembly systems)

.

I I I I I

V. SEQUENCE NUMBER (for use in assembly systems and for I I

10 I I I I I I

those who prefer to use a straight sequence number rather I , I

than a page and line reference) I 1 I

11 I I I ,

.

I I , I I I

I I j I I

12 VI. PAGE NUMBER (used to designate the page where reference , I I I I

I I I I I

I I , I ,

is made to this instruction) I I I I I

13 I I I I I

, I , I I I I I I

VII. LINE NUMBER (used with Page Number to designate a specific I I I

14 I I I

line on which reference is made to this instruction) I , I , I ,

I I ,

15 I , I , I I ,

I , ,

r-- I I I I I I , I

16 I I I I I , I I I I I I I I I I I I I I I

f - -

I I I

I I I

17 I I I I I I I

I I I , I

I I , ,

I-- I I I , I I

18 I I I I I I I I I I I I I

I I I I I I I

I - -

I I I I I I I I I

I 19 I I I I I I I I I I

I I I I I I I I I I I I I

I I I , , I I I , , I I

I I--

I I I I I I I I I I I I I I

20 I , I I I I I I I I I , I I

I I I , , I I I I I I ,

I I I I I I I I I I I , , ,

I

' - -

KEY: t. TRANSLATE AND STORE + 2. TRANSLATE AND STORE - 3. STORE UNPRIMED + .t. STORE UNPRIMED - 5. STORE PRIMED+ 6. STORE PRIMED _

(5)

~ .

~ SOLlD.STATE COMPUTER· CODING CHART

.. R.III....

~D~ TItAININ. D."A~ItTl"NT

DIYISION Of spun lAND CORPQRA·TlON

ICARD

I

ROUTINE _r.--- ... " I."'.,. r-- ,.."U"'"~u .. ; ""\lie:.: I

---

REFERENCE INSTRUCTION K

/

SEQUENCE LOCATION WORD E NOTES

TIMe:

PAGE ILINE X a Op I

m

I

c

y 'rA rX

.

rL/

!

1

@ @i ® !@ H

I I I I I I I I , I I I I , I. I I I I I I , I I I I I I , , I I I , I I I I I I , , I I , I , I , , , , I I I , , , , I I I I , I I , , I

,

I I I " I I

Yr'

I I I , I I , , , I I ,

:

, ~ I

.

I , .. I I

, 2

""

, I / l

T

I I

, I I

, I I

"l

3

,

u

.-!_~_~~~~l~~_~!!~~

2

__ Jl--".'-~~J_Ll~-* [b~!· L_".(:1~J~!h~J~~!! ~~~~t---.---

REFERENCE INSTRUCTION

--,

Z I I I I I

""""

, ... I

, ,

4 i 34 _ T 34 I

~ ,

I~~

WORD LJ

I

.~ 5

" ... ®... "

, ,

78

I~l D ® F G

78 , , ,

.}!

,

a m c I I I , , ,

I • • • 5 7 • • ~ "i~ u U U I " U " ~" u " " H H.V H "~ ~ U P M • • " . ~ u a ~ Q I I

7 Q

---~---~---

I I I

z I , I

.., I I I

II: I I I

Z i : I

8 0

I

...

,

"

12 12 I I I

Z I I ,

i I I I

9 ~ 34 UNIVAC SOLID·STATE 90 34 I , ,

I I

.(

,

10 <Ii S6 56 i I

PROGRAM CARD

,

:::) I

.: 78 78 I

11 'C " I I

£ 9 9 , ,

II: ••• 7 Q •• S O I I 5 2 5 3 M 15 11 17 51 " m U R a l l • • l7u.~nn-UBH"nHmuRall.lIl7 • • ~ ,

12 I I

I I I I I I

..

,

.

I I

13

-

I I I I I

,

I I I I I 1 I I I I I I I I I I I ..

I I I I I

,

I I I I I I I I I I I I I I I I I I

I I I I I I I I

I , I I I I I I I I I I I I I I

! ! ! ~ ! I ~ ! , I

.

! : : I I

. .

I I I : ! ! " : ! !

14 Note:

-

Some programmers may prefer to use a straight Sequence , 15

I number rather than a Page and Line reference. In this

i

I

-

event the actual Page (A) and preprinted Line (B) numbers

I 16

I I

are ignored and the Sequence number is written in the

I r--

I I

17 reference area under the captions Page and Line. Sequence

,

I number is punched in card columns 11 through 15.

I ~I

I I

18

I I

,

I t---

I 19 I I I I I I I , I I I I I I I I I I I I I I I , I I I

I I I I I I I I , I

I I I I I I I I I I I I I I I I I I

I I I , , I , I I I

I , , I I I I I I I I I I I I I I I

I I I I I I I I I I I I I I I I I I I ~ I I I I

I-- I I

I I I I I I I I I I I I I I , I I I , I I I I I I , I I

I 20 I I I I I I I I , I I I I I I I I I I I I I I I I I I

I I I I I I I , I , I I I I I I I I , I I I I I I I I

I , I , I I I I , I I I I I I I , I , I I I I I I I I

L . -

KEY: 1. TRANSLATE AND STORE + 2. TRANSLATE AND STORE_ 3. STORE UNPRIMED + 4. STORE UNPRIMED - 5. STORE PRIMED+ 6. STORE PRIMED- U1711 R .... l

(6)

NEW UNIVAC@MAGNETIC AMPLIFIER, SOLID.STATE COMPUTER. MEMORY CHART

CUSTOMER: __________________________ ___ PROGRAM: ________________________________ __

APPLICA TI ON : PAGE :

BAND: BAND:

__ 00 TO __ 99 __ 00 TO _ _ 99 __ 00 TO -_99 - _00 TO __ 99

00 50 00 50 00 50 00 50

01 51 01 51 01 51 01 51

02 52 02 52 02 52 02 52

03 53 03 53 03 53 03 53

04 54 04 54 04 54 04 54

05 55 05 55 05 55 05 55

06 56 06 56 06 56 06 56

07 57 07 57 07 57 07 57

0.8 58 08 58 08 58 08 58

09 59 09 59 09 59 09 59

10 60 10 60 10 60 10 60

11 61 11 61 11 61 11 61

12 62 12 62 12 62 12 62

13 63 13 63 13 63 13 63

14 64 14 64 14 64 14 64

15 65 65

16 66 Each time an Instruction or Constant is assigned a memory 66 17 67 location write the reference, either Page/Line or Sequence 67

18 68 68

19 69 number, opposite the word designation on the memory chart.

This will eliminate the possibility of assigning a memory 69 20 70 location to more than one instruction/constant and aid the

70 21 71 programmer in selecting unused locations.

71

22 72 72

23 73 23 73 2::1 1:3 23 73

24 74 24 74 24 74 24 74

25 75 25 75 25 75 25 75

26 76 26 76 26 76 26 76

27 77 27 77 27 77 27 77

28 78 28 78 28 78 28 78

29 79 29 79 29 79 29 79

30 80 30 80 30 80 30 80

31 81 31 81 31 81 31 81

32 82 32 82 32 82 32 82

33 83 33 83 33 83 33 83

34 84 34 84 34 84 34 84

35 85 35 85 35 85 35 85

36 86 36 86 36 86 36 86

37 87 37 87 37 87 37 87

38 88 38 88 38 88 38 88

39 89 39 89 39 89 39 89

40 90 40 90 40 90 40 90

41 91 41 91 41 91 41 91

42 92 42 92 42 92 42 92

43 93 43 93 43 93 43 93

44 94 44 94 44 94 44 94

45 95 45 95 45 95 45 95

46 96 46 96 46 96 46 96

47 97 47 97 47 97 47 97

48 98 48 98 148 98 48 98

4 49 99 49 99 49 99 49 99

U 1710

(7)

B. To facilitate automatic sorting of instruction and constant words it is suggested that blocks of Page and Line or Sequence numbers be assigned as follows:

1. A major block for all instruction words.

2. A minor block for all constants which are instruction words.

3. A second minor block for all constants which are not instruction words.

C. Consideration should be given to the advantages of punching a one (I) in some unique position of all program cards to estab- ish a card count control.

D. To aid in the automatic analysis it will be helpful to prepare and verify Program Testing Decks. (Note that Band I, words 0000 through 0199, is presently reserved for control routines) Three program iesting decks will be required:

I.

2.

3.

III. PROCEDURES

PT-I - 4800 cards with 4999 punched in 17 - 20.

PT-2 - 4800 cards with 4999 punched in 23 - 26.

PT-3 - 4800 cards with 4999 punched in 27 - 30.

memory locations 0200 through the "a" location, card columns

memory locations 0200 through the "m" location, card columns

memory locations 0200 through the "c" location, card col umns

A. Verify Key Punching

I. List all the Program Cards on the tabulator in the same order as the coding appears on the coding charts. Set sub-total control on Pagp number (columns II - 13) to obtain space between pages.

2. Verify to original coding charts. Save this list and manually post all corrections made as a result of sub-

sequent analysis.

5

(8)

6

B. Verify legality of "a" addresses in all program cards.

1. Sort all program cards, instructions/constants, to "a"

address, card columns 17 - 20.

2. Merge program testing deck # 1 into the program deck.

Segregate duplicates and non-matching cards from the program deck. A 315 or 319 Collator will:

\

(a) Segregate program cards if more than one in- struction/constant was assigned to a memory location.

(b) Segregate invalid "a" addresses.

(c) Interfile a blank "a" address card into the program deck for each location not used.

3. List on tabulator, print all data. This listing will serve to replace Memory Chart (U1710) by providing complete information relative to the line of coding and provide a space opposite unused words for noting addi- tional instructions, if necessary.

C. Verify legality of Operation Codes

1. Sort instruction and constant cards which represent in- structions to Operation Code, card columns 21 - 22.

2. List on tabulator, print all data. Set sub-total con- trols on Operation Code to obtain spacing between codes.

(a) Scan list visually for illegal Operation Codes.

(b) Check Operation Codes which should be blank in

"m" or "c" locations.

(c) Check for correct stacker location in 47 and 57 codes.

(d) In the 81,96,11 and 46 codes check to be sure that the "c" address is no earlier than the re- commended location for minimum latency.

(e) In the same 81,96,11 and 46 codes check to be sure that the instruction address is no later than that recommended for minimum latency.

(f) Scan shift codes for correct location of "n"

(9)

(g) Optional: if word times have been punched in the cards and if a 330 Calculating Punch,

UNIVAC 60/120 or UNIVAC SOlid-State 90 is avail able, calculations may be made to determine the word times between instruction address and the

"c" address to check the word times determined by the programmer.

(h) If a UNIVAC 60/120 or UNIVAC Solid-State 90 is available, it may also be desirable to program to determine whether the "c" is spaced from the

"m" address by at least the execution word times.

D. Verify legality of '~" addresses in instruction cards

1. Sort instruction and constant cards which represent in- structions to Page and Line, card columns 11 - 15,

(minor) then to '~" address, card col umns 23 - 26 (major).

2. Merge program testing deck U 2 into the program deck.

Segregate non-matching cards from the program deck. A 315 or 319 Collator will:

(a) Segregate invalid "m" addresses.

(b) Interfile a blank '~" address card into the program deck for each memory location not used as "m".

3. List on tabulator, print all data. Set sub-total con- trols on two most significant digits of "m" to obtain spacing. This listing will group all '~" references to memory by Page and Line or Sequence number and provide a single blank space opposite each memory location not used.

4. List on tabulator all program cards which were segregated on the merge operation. These cards will be:

register addresses shifts of 10 positions addresses in machine code.

(a) Check carefully all instances where the same "m"

address is in more than one card. Duplicates may be for legal multiple use of the same op-

erand or may indicate erroneous coding. Use Page and Line or Sequence number to refer to

coding charts.

7

(10)

8

(b} Check "m" addresses in interlace positions to be sure that correct addresses have been used for first read, second read, print and punch interlace words.

E. Verify legality of "c" addresses in instruction cards

1. Sort instruction and constant cards which represent in- structions to Page and Line, card columns 11 - 55,

(minor) then to "c" address, card col umns 27 - 30 (major).

2. Merge program testing deck # 3 into the program deck.

Segregate non~atching cards from the program d&ck. A 315 or 319 Collator will:

(a) Segregate invalid "c" addresses.

(b) Interfile a blank "c" address card into the pro- gram deck for each memory location not used as

"c" •

3. List on tabulator, print all data. Set sub-total con- trols on the two most significant digits of "c" to obtain spacing. This listing will group all "c" references to memory by Page and Line or Sequence number and provide a

single blank space opposite each memory location not used.

4. List on tabulator all program cards which were segregated on the merge operation. These cards will be register addresses.

5. Check carefully all instances where the same "c" address is in more than one card. Duplicates may indicate a legal condition, as where the "c" address concerned is the point to which a number of instructions jump, or may indicate erroneous use of same "c"address in two or more instructions. Use Page and Line or Sequence numbers

to refer to coding charts.

6. Check "c" addresses for all words which are in bands used as interlaces in the program to insure that the correct location has been used for first read, second read, print and punch data. For example, if an instruction refers to a "c" address which is supposed to be in first read words, determine that the address actually is in first read inter- lace position.

(11)

F. Verify continuity from "c" address to "a" address

1. Remove all PT-3 cards from program cards used in the "c"

verification above.

2. Keeping the program cards in "c" address order,reproduce the instruction cards gang punching into the reproduced cards a carriage line space control hole and some print- able position to identify each card as being part of the reproduced deck.

3. Sort reproduced deck (PT-4) on "a" address, card columns 17 - 20.

4. On a 319 Collator merge the program cards ahead of matching cards in the reproduced deck. Compare the "c"

address in the program deck, card columns 27 - 30 to the

"a" address in the reproduced deck, card col umns 17 - 20.

Segregate non-matching cards from both decks.

5. List all cards on the tabulator, print all data. Set carriage line spacing to 1/1.

6. For nearly every address "c" of a true instruction card there should be a matching instruction address, ("aft), in the reproduced card listed on the line below. The carriage line space control hole separates the related pair for reading convenience.

(a) Check all cases where a single line appears for either the original deck or reproduced deck.

Be sure to print the identification of the re- produced deck.

(b) Single lines may indicate constants, or may indicate a "c ft address reference to an instruc- tion address which is not represented by a card, or may indicate a register which is used to contain the next instruction, or may indicate one of the addresses involved in a comparison operation.

G. Verify insertion of 22 test of HSR buffer

1. Return to original coding charts. Be sure that all changes from preceding procedures have been made on the chart and in the program cards.

9

(12)

10

2. For each type of input card and every branching condition possible from the card, add up the word times on a

Remington adding machine or "99" calculator from each 42 test in the instruction sequence to the next applicable 42 test. Check accumulated word times to insure that the

interval from one 42 test (including executing that test) up to the next 42 test does not exceed 1000 word times

IV. CONSIDERATIONS

(5 drum revolutions). Save the printed copy to file with the coding papers, identified by input condition tested.

Note that the instruction address can be entered at the left of the word times, allowing four total positions for the word times, thus providing reference to the instruc- tion address related to the word times figure.

A. When frequent use is made of the standard High-Speed Reader Reserve Storage Routine and Read-Punch Unit Backup Punching Routine, it may be desirable to include the complete coding in the three program testing decks.

B. Some programmers will find it advantageous to punch a designa- ting position into each of the program testing decks to iden- tify each memory location used in an interlace.

C. It will always be Gesirable to rerun each of the lists de- scribed above when final corrections have been made. These lists should become part of the Run Book. Corrected lists will insure efficient use of time when debugging at the computer.

(13)

...

~" ••

.

UN I V A C L Tha FIRST Name in I!la(lltonio CompU11ntl Syslems

Références

Documents relatifs

These functions include the controlled generation of vertical Bloch lines in a thin (111)- oriented garnet film with uniaxial anisotropy,the propagation of VBL- pairs having

Definition. The whole set of formal concepts thus corresponds to all the possible maximal correspondences between a set of objects O and a set of attributes I. Indeed, the user

In this study, our objectives were to investi- gate on-line and off-line physical measurement with a new device during axenic cell cultures under high biomass concentration.

In this paper, we propose a non-invasive method to determine the electrode balancing of the lithium-ion batteries, which is the determination of (i) individual electrodes capacities

Under the one-port model, with a uniform communi- cation time for all tasks and processors, the optimal policy of the master consists in sending the tasks as soon as they arrive..

As the on-line recognizer, we employ a structured character pattern representation (SCPR) dictionary to reduce the total size of memory and a Linear-time Elastic Matching (LTM)

KEY WORDS: mobile mapping system, photogrammetry, multi-camera panoramic head, calibration, off-line calibration, on-line calibration, panoramic head calibration with very low

Table 7 shows the number of instances with the minimum calculation time required to obtain a proved optimal solution ( Best − time ) when both models guarantee the optimal solution